BludgerTrack records a movement to the Coalition this week, but in keeping with the zeitgeist, this is more about changes in preference assumptions rather than voting intention. Specifically, I’ve decided to apply a crude 60-40 split in favour of the Coalition on One Nation preferences, as Newspoll has been doing since the start of last year.
A while back I came up with an elaborate mechanism to allocate One Nation preferences based on respondent-allocated two-party polling data, the true purpose of which was to produce a figure more favourable to the Coalition than the 50-50 split recorded in the 15 seats the party contested in 2016, which only partisan optimists (hello to you all) expect to be repeated this time. However, this has been increasingly ineffective due to the paucity of respondent-allocated results since ReachTEL’s national polling stopped around a year ago. It seemed to me that something needed to be done though, and I have been persuaded by the position of David Briggs at YouGov Galaxy that 60-40 is a conservative approximation (albeit an arbitrary one) given the preference flows at the last two state elections at which One Nation made a serious effort in lower house seats, namely Queensland (65.2% of preferences to the Liberal National Party) and Western Australia (60.6% to the Liberals).
I am not, however, convinced that the same thing should be done with the United Australia Party, as Newspoll has now started doing. The Palmer United Party had Labor last on every how-to-vote card in 2013, yet 46.3% of their voters still put Labor ahead of the Coalition. In addition to the impact of the heavily publicised preference deal, Briggs points to the fact that UAP voters in the latest Newspoll sample strongly favoured Scott Morrison over Bill Shorten on the question of trust, but this strikes me as thin gruel given the small sample size. Kevin Bonham makes the point that the Sinophobic bent of Palmer’s current campaign might be capturing a more right-wing audience than last time, which may well be so. However, he also makes the very good point that Palmer “may be taking Coalition-friendly voters from the Others pile, so the remaining Others may on balance be slightly Labor-leaning”.
All things considered, I don’t see enough reason to stop treating the UAP as part of the amorphous collection of “others” and to continue allocating its collective preferences as per the 2016 result, which was basically 50-50 – particularly not in the context of an election at which anti-government sentiment is harder than it was last time, based on all available evidence. In any case, I will not for the time being be making the effort to produce a trend measure from the UAP, whose primary vote will remain locked up in BludgerTrack’s aggregated “others” measure.
The upshot of all this is that the dial has moved 0.5% in favour of the Coalition on two-party preferred, but only 0.2% of this is due to the addition of the new polls this week from YouGov Galaxy, Newspoll and Essential Research. The Coalition has gained three on the seat projection, consisting of one each in Queensland, Western Australia and South Australia. The addition of new state data has smoothed off what hitherto seemed excessive movement in the Coalition’s favour in New South Wales, although it’s had the opposite effect in Western Australia. Labor continues to be credited with eight gains in Queensland, which seems rather a lot, but elsewhere the projections seem in line with what the major parties are expecting.
Full results can be accessed through the link below, which is permanently available on the sidebar.
And while you’re about, don’t miss the latest edition of Seat du Jour in the post below this one, covering Peter Dutton’s seat of Dickson.